Aerial Wilderness Search and Rescue with Ground Support

被引:33
作者
Kashino, Zendai [1 ]
Nejat, Goldie [1 ]
Benhabib, Beno [1 ]
机构
[1] Univ Toronto, 5 Kings Coll Rd, Toronto, ON M5S 3G8, Canada
基金
加拿大自然科学与工程研究理事会;
关键词
Autonomous mobile-target search; UAV-UGV Cooperative search planning; Iso-probability curves; Wilderness search and rescue; ROBOTIC INTERCEPTION; TARGET DETECTION; MOVING-OBJECTS; UAV; SURVEILLANCE; SYSTEM; AIR; COVERAGE; TRACKING; STRATEGY;
D O I
10.1007/s10846-019-01105-y
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Unmanned aerial vehicles (UAVs) have been proposed for a wide range of applications. Their use in wilderness search and rescue (WiSAR), in particular, has been investigated for fast search-area coverage from a high vantage point. The probability of success in such searches, however, can be further improved utilizing cooperative systems that employ both UAVs and unmanned ground vehicles (UGVs). In this paper, we present a new coordinated-search planning method, for collaborative UAV-UGV teams. The proposed method, particularly developed for WiSAR, considers the search area to be continuously growing and that the search is sparse. It is also assumed that targets detected by UAVs must be identified by a ground-level searcher. The UAV/UGV motion-planning method presented herein, therefore, has two major components: (i) coordinated search and (ii) joint target identification. The novelty of the proposed method lies in its use of (i) time-dependent target-location iso-probability curves, and (ii) an effective and efficient coordinated target-identification algorithm. The method has been validated via numerous simulated WiSAR searches for varying scenarios. Furthermore, extensive comparative experiments with other methods have shown that our method has higher rates of target detection and shorter search times, significantly outperforming alternative techniques by 75% - 255% in terms of target detection probability.
引用
收藏
页码:147 / 163
页数:17
相关论文
共 65 条
  • [1] Agcayazi MT, 2016, INT CONF UNMAN AIRCR, P898, DOI 10.1109/ICUAS.2016.7502618
  • [2] [Anonymous], [No title captured]
  • [3] [Anonymous], P IEEE S SERIES COMP, DOI DOI 10.1109/SSCI.2017.8280972
  • [4] Decentralized planning and control for UAV-UGV cooperative teams
    Arbanas, Barbara
    Ivanovic, Antun
    Car, Marko
    Orsag, Matko
    Petrovic, Tamara
    Bogdan, Stjepan
    [J]. AUTONOMOUS ROBOTS, 2018, 42 (08) : 1601 - 1618
  • [5] An active vision system for multitarget surveillance in dynamic environments
    Bakhtari, Ardevan
    Benhabib, Beno
    [J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2007, 37 (01): : 190 - 198
  • [6] Active-vision-based multisensor surveillance - An implementation
    Bakhtari, Ardevan
    Naish, Michael D.
    Eskandari, Maryam
    Croft, Elizabeth A.
    Benhabib, Beno
    [J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART C-APPLICATIONS AND REVIEWS, 2006, 36 (05): : 668 - 680
  • [7] Beck Z, 2016, AAMAS'16: PROCEEDINGS OF THE 2016 INTERNATIONAL CONFERENCE ON AUTONOMOUS AGENTS & MULTIAGENT SYSTEMS, P1024
  • [8] Navigation-guidance-based robotic interception of moving objects in industrial settings
    Borg, JM
    Mehrandezh, M
    Fenton, RG
    Benhabib, B
    [J]. JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS, 2002, 33 (01) : 1 - 23
  • [9] Brown D, 2017, INT CONF UNMAN AIRCR, P1425
  • [10] Caska S., 2014, CIE44&IMSS, P453